# Status page

A self-hosted status page with incident management for SaaS products and internal services, built with Next.js, TypeScript and PostgreSQL. It covers grouped components with 90-day uptime bars, incidents with updates and write-ups, planned maintenance, history by month, Atom and RSS feeds, and optional HTTP checks run from a scheduler.

- Framework: Next.js (JavaScript)
- Category: Apps / Developer tools
- Creates: a server app and a PostgreSQL database on the Shared tier (1 GB)
- Plan: Lite or above
- Price: Free
- Author: Light Cloud
- Page: https://templates.light-cloud.com/apps/dev-tools/status-page-template
- Updated: 2026-11-21

## Features

- Components in groups with five statuses, from Operational to Major outage and Under maintenance, each shown with a word and an icon, never colour alone
- 90 daily uptime bars per component with its uptime percentage, where you pick a day with the mouse, a tap or the arrow keys to see what happened
- Incidents with a stage, an impact, affected components and a timeline of updates; resolving puts every affected component back to Operational
- Planned maintenance windows that show components as Under maintenance and never count against uptime
- Markdown incident write-ups, history by month, a URL per incident that stays the same when you edit the title, and Atom and RSS feeds
- Optional HTTP checks that GET or HEAD a URL and mark a component down after a set number of failures in a row, run when a scheduler calls a secret check URL
- A team app with owner, admin and member roles: an overview with a status picker per component, editable updates and components ordered with move up and down
- A demo Invoicing App page with 11 components, an open incident and 7 past incidents, removed in one click, and WCAG 2.2 AA in light and dark

## What you need

- A paid plan: deploying creates a PostgreSQL database, which the Free plan does not include
- To run it locally: Node.js 22.12 or newer, npm, and Docker for a local PostgreSQL
- Optional: a scheduler or uptime pinger that calls /api/cron/check every 1 to 5 minutes; without one the HTTP checks never run
- Optional: CRON_SECRET for the check URL; without it a secret derived from AUTH_SECRET is used and shown on the Checks screen
- Optional: RESEND_API_KEY and EMAIL_FROM to email invitations and password resets, and GitHub or Google keys for team sign-in

## About

The page's name, description and links are edited in the app under Settings, Status page, components and groups in the team app, the time zone with TZ, and the colours, including the five status colours, at the top of src/app/globals.css; the demo Invoicing App page is in src/lib/demo. Every status change, by hand, from an incident or from a check, writes a row that the uptime bars read and split into calendar days in TZ. HTTP checks run only when a scheduler or uptime pinger calls /api/cron/check with the secret, so this is not continuous monitoring, and each call wakes the app and uses compute time. Without CRON_SECRET the check secret is derived from AUTH_SECRET, so rotating AUTH_SECRET breaks scheduled calls until they are updated. Owners and admins can point a check at any URL and the app fetches it, keeping only the status code and the time taken. There are no email or SMS subscriptions, maintenance windows don't start on their own, and a failed check raises an alert on the overview rather than opening an incident; host the page apart from the product it reports on, so it stays up when the product is down.

## Variables

- `AUTH_SECRET`: generated

## Questions

### Is there a free open-source status page I can self-host?

Yes. This status page is MIT licensed and built with Next.js 16, PostgreSQL, Drizzle ORM and Better Auth. It runs with no third-party keys and starts with a demo page you remove under Settings, Status page.

### Does this status page monitor my site automatically?

Only when something calls it. The HTTP checks run each time a scheduler or uptime pinger calls /api/cron/check with the secret, typically every 1 to 5 minutes, and a component is marked down after the number of failures in a row you choose. Between calls nothing runs, because the app can scale to zero.

### How is uptime calculated on the status page?

Each bar is one calendar day in your time zone. Uptime is the share of the last 90 days a component was not in a partial or major outage; degraded performance counts as up and planned maintenance is left out.

### How do customers get notified about incidents?

They can follow the Atom feed at /feed.atom or the RSS feed at /feed.rss in a feed reader or a chat tool that reads feeds, where every new incident, update and maintenance window appears. Email and SMS subscriptions are not built in.

### Can I host this status page without Light Cloud?

Yes. It is a standard Next.js app with no Light Cloud code, so any host with Node.js 22.12 and PostgreSQL runs it with npm ci, npm run build and npm start once DATABASE_URL and AUTH_SECRET are set.
